FEI's Peg Break Tests the Limits of Algorithmic Stablecoin Design

FEI launched with backing from Andreessen Horowitz, Framework, Coinbase, and other prominent names — a lineup that generated substantial hype but ultimately proved no guarantee of success in DeFi. Big-name investors are accustomed to buying at favorable valuations, but their involvement doesn't automatically translate into the value the market often assigns it.

FEI lost its $1 peg, and TRIBE, the protocol's governance token, fell roughly 30% from its launch price. The mechanics of Fei's tokenomics ended up penalizing users for problems created by the project's own design choices, and the launch is widely regarded as a failure. The episode raises a broader question about where algorithmic stablecoins go from here.

Background: two competing models

Two academic papers published in 2014 laid out the conceptual groundwork that most algorithmic stablecoins still draw on today. Ferdinando M. Ametrano's Hayek Money introduced a rebasing approach, where total supply adjusts with demand. Robert Sams' Seigniorage Shares proposed a two-token structure — a stable asset paired with a volatile "seigniorage share" token whose holders absorb the gains and losses tied to supply and demand shifts.

These two frameworks underpin later projects such as ESD, Basis, RSR, and Frax, and now FEI as well.

Collateral-backed stablecoins remain the closest thing to a proven model, but the decentralized versions of that approach are capital-inefficient, expose the backing assets to risk, and see limited adoption because of the overcollateralization they require. Ampleforth and Empty Set Dollar tried to sidestep collateral entirely, but neither delivered real price stability. Maker's DAI is generally considered the strongest decentralized attempt, relying on user-supplied collateral, yet even it isn't frictionless — closed-loop arbitrage for DAI is not achievable, which keeps its price from being perfectly stable and creates scaling constraints.

Total stablecoin supply climbed more than 577% between the start of 2020 and the start of 2021, going from $5.9 billion to over $40 billion. FEI arrived as part of what's been called the second wave of algorithmic stablecoins, alongside ESD 1.5 and cross-chain designs like Frax. Despite a rocky launch that dented confidence in the project, the broader search for a workable "immaculate stablecoin" continues to draw builders.

How FEI was supposed to work

FEI's design combined a bonding curve with a "direct incentives" mechanism intended to:

maintain a liquid market in which ETH/FEI trades at approximately the ETH/USD price.

During the widely anticipated Token Genesis Event, participants could mint FEI from the ETH bonding curve at discounts of up to 50%, while also qualifying for a TRIBE governance token airdrop. The offer drew more than 17,000 unique addresses.

That discount turned out to be the first disappointment in a longer string of them. Its terms capped eligibility at $250 million in contributed ETH — but the launch pulled in $1.3 billion instead, voiding the discount and sending all of that ETH irreversibly into the project's "Protocol Controlled Value" reserve, described as giving:

the protocol more flexibility to engage in activities that are not profit-oriented.

That entire PCV pool was initially deployed to seed liquidity on Uniswap, and at its peak the FEI/ETH Uniswap pool held over $2.6 billion. None of that liquidity depth mattered much once FEI's peg — and the project's credibility along with it — began to slip.

What broke the peg

The initial de-pegging traced largely to a "pre-swap" feature that let genesis participants automatically convert their FEI into TRIBE at launch, generating heavy sell pressure on FEI almost immediately.

Fei's mechanism penalizes FEI sellers and rewards buyers whenever the token trades below peg. Because roughly 98% of TRIBE's liquidity sits in the TRIBE-FEI pair, anyone trying to exit TRIBE for ETH was stuck choosing between two bad options: route through TRIBE-FEI and eat the below-peg selling penalty, or accept heavy slippage from the thin TRIBE-ETH pair.

Compounding matters, a "vulnerability was found in the incentive calculation," despite the protocol having been audited by both Consensys Diligence and Open Zeppelin. The team's response was to disable all minting rewards, noting that:

reweights will still occur without mint rewards.

A chart circulated by @bantg illustrated how these reweighting events affected the price.

Economist and researcher Emin Gün Sirer laid out the mechanism design flaw on Twitter:

Let's examine FEI. The thinking goes like this: Why do stablecoins go lower than their peg? Because people sell. So let's punish people who sell FEI when it's trading below the peg!

The lower the price, the greater the punishment, aka penalty. If you sell FEI when it's trading below its peg, the recipient receives fewer FEI than you transferred, taking some of the transferred coins out of circulation. The thinking is that this should boost the price. But now, instead of just doing the obvious, single step of logical inference ("let's penalize transfers when below the peg"), let's look ahead a bit.

Every sale has a buyer and a seller. If you punish the sellers, you're also punishing the buyers.

The penalty mechanism in FEI not only makes the supply disappear, it also makes demand disappear.

It punishes both sides, and thereby narrows the feasibility envelope for the coin.

Making the effective price lower when the price is low is exactly the opposite of what you want in an algorithmic stablecoin!

Fallout and reactions

When one of Fei's founders floated the idea of putting some of the PCV to work in Yearn, the suggestion drew mockery on Twitter, with Karoo quipping:

Our project isn't quite going as planned, so why don't we at least get some yield on all this eth you gave us.

Banteg, posting in a public group, took a harder line:

Personally I'd vehemently want them to dismantle and refund their users, even if it means not deploying +50% tvl to Yearn.

We don't decide who uses the protocol. After all, neutrality is something we all have been fighting for.

Julien Bouteloup pushed back against that framing of neutrality, publicly posting a list of FEI holders he argued should be excluded from future airdrops.

Reactions to the whole episode split sharply: some called it a failed experiment deserving a full ETH refund to depositors, while others doubled down, publicly pledging to buy unlimited FEI as a show of confidence — as in this comment from R. Leshner:

I'm still $0.70 USDC bid. I'll buy all the $FEI you can sell me there.

Where that leaves things

A middle position is probably the more defensible one. The project, by most measures, hasn't succeeded so far — but it was framed from the outset as an experiment, and experiments sometimes fail. Participants had access to all the relevant information before depositing, and a blanket expectation of refunds for every failed DeFi experiment would likely chill the willingness to try new designs at all.

At the same time, there was no real necessity to raise the sums Fei did, and it's fair to ask why backers would commit that much capital to founders with a track record of struggles on smaller projects. Prominent figures publicly staking their reputations to the token's price — and nudging others toward buying at or above current levels — didn't help matters either.

As Fiskantes put it on Twitter:

The most crowded investment opportunities are NEVER the best ones.

Whether financial experiments of this scale keep attracting this level of funding, or whether FEI marks something of an inflection point, remains an open question.